Data mining in E-commerce - building anime recommendation system

Abstract

This research aims to build an anime movie recommendation system based on the nearest neighbor method, one of the recommendation methods based on users with similar behavior, to provide users using an objective system that is simple and easy to use. Because our knowledge is still quite limited, we have consulted a number of other recommendation systems and their results will be clearly stated in the research article. The result was research and development using the Anime-Planet dataset which includes information on 16,621 different anime, from 74,129 users and 16,000 animes. Some information about tags, ratings, genres, reviews, viewing status, viewed status,.... and the data set is divided into many different small files, we have come up with a recommendation system. recommend anime to users based on nearest neighbor method, using streamlit, pandas, NumPy libraries.